%% Parameters

% Quality grid    
    param.Q 	= 100;  % # of q grids
    param.d 	= 5;    % quintiles
    param.lb    = 0; 	% lower bound of q grid
    param.ub    = 8;	% upper bound of q grid      
   
% Omega grid    
    param.n1	= 10000;            % # of omega1 grids
    param.n0 	= 5;                % # of omega0 grids
    param.Nf	= param.n1*param.n0;% total number of firms (types)  
    param.omega0_mean   = 0;    % mean of omega0 normal distribution   
    param.omega1_mean   = 0;    % mean of omega1 normal distribution  

% Export demand shock
    param.shock = 1.05;
        
% Normalization
    param.X     = 1;    % output
    param.P_s   = 1;    % service price
    param.e     = 1; 	% exchange rate
    param.w     = 1;  	% efficiency wage    
    
% Parameters         
    param.sigma = 5;  
    param.sigma_s = 5;      
    param.f_vH  = 1;	% seller search cost at home
    param.f_vF  = 1;	% seller search cost at foreign
    param.f_m   = 1;	% buyer search cost
       
% Calibrated from Table 1 of "Empirics/Results/14-October2019/Notes.pdf"
    param.alpha_m   = 0.33;  % manufacturing input
    param.alpha_s   = 0.38;  % service (1-0.33-0.29=0.38)
    
% Calibrated from Table 6 of March 2020 Draft
    param.beta_v    = 1/0.462; 
    param.beta_m    = 1/0.593;      
    
% QQ regressions [all, top50, top75, top95, top96, top97, top98, top99]
    param.qq_slope = [0.290, 0.495, 0.609, 0.605, 0.605, 0.603, 0.601, 0.564];
    param.qq_const = [0.000, -0.195, -0.357, 0.349, 0.349, 0.345, 0.338, 0.234];    
    param.wage_mu = param.qq_const.*param.qq_slope;
    param.wage_sigma = param.qq_slope;  
    
    